The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface because it ingests untrusted data from external sources. Ingestion points: In scripts/agent.py, the test_ssrf_payload function fetches responses from external target web applications via requests.get and records a preview of the response body. Boundary markers: There are no explicit boundaries or instruction delimiters implemented to isolate the ingested response content if downstream LLM processing occurs. Capability inventory: The script performs outbound network probes and writes data to a local report file. Sanitization: The captured external text is included raw in the report without sanitization or escaping, creating a pathway for malicious content from a target site to be reviewed by the agent.

performing-ssrf-vulnerability-exploitation